No abstract

A gravity theory is developed with the metricĝ µν = g µν +B∂ µ φ∂ ν φ. In the present universe the additional contribution from the scalar field in the metricĝ µν can generate an acceleration in the expansion of the universe, without negative pressure and with a zero cosmological constant. In this theory, gravitational waves will propagate at a different speed from non-gravitational waves. It is suggested that gravitational wave experiments could test this observational signature.

Abstract. The dynamical consequences of a bimetric scalar-tensor theory of gravity with a dynamical light speed are investigated in a cosmological setting. The model consists of a minimallycoupled self-gravitating scalar field coupled to ordinary matter fields in the standard way through the metric: gµν +B∂µφ∂νφ. We show that in a universe with matter that has a radiation-dominated equation of state, the model allows solutions with a de Sitter phase that provides sufficient inflation to solve the horizon and flatness problems. This behaviour is achieved without the addition of a potential for the scalar field, and is shown to be largely independent of its introduction. We therefore have a model that is fundamentally different than the potential-dominated, slowly-rolling scalar field of the standard models inflationary cosmology. The speed of gravitational wave propagation is predicted to be significantly different from the speed of matter waves and photon propagation in the early universe.

The arguments leading to the introduction of the massive Nonsymmetric Gravitational action are reviewed [1,2], leading to an action that gives asymptotically well-behaved perturbations on GR backgrounds. Through the analysis of spherically symmetric perturbations about GR (Schwarzschild) and NGT (Wyman-type) static backgrounds, it is shown that spherically symmetric systems are not guaranteed to be static, and hence Birkhoff's theorem is not valid in NGT. This implies that in general one must consider time dependent exteriors when looking at spherically symmetric systems in NGT. For the surviving monopole mode considered here there is no energy flux as it is short ranged by construction. Further work on the spherically symmetric case will be motivated through a discussion of the possibility that there remain additional modes that do not show up in weak field situations, but nonetheless exist in the full theory and may again result in bad global asymptotics. A presentation of the action and field equations in a general frame is given in the course of the paper, providing an alternative approach to dealing with the algebraic complications inherent in NGT, as well as offering a more general framework for discussing the physics of the antisymmetric sector.

The massive nonsymmetric gravitational theory is shown to possess a linearization instability at purely GR field configurations, disallowing the use of the linear approximation in these situations. It is also shown that arbitrarily small antisymmetric sector Cauchy data leads to singular evolution unless an ad hoc condition is imposed on the initial data hypersurface.